1

High Attitude Student Platform

Inter-American University, Bayamon Science Report

Team Name: ARIES-Dynamics Team. Payload #6

Team Name

Cover Page

Prepared by: Dr. Hien B. Vo

Project Principal Investigator Date

Prof. Diego Aponte

Advisor/ Mentor Date

Damian Miralles Sanchez

Team Member Date

Jean P. Ojeda Garcia

Team Member Date

Nelson Nieves

Team Member Date

Francisco Franquiz

Team Member

Luis

Team Member Date

2

Table of Contents 1

Cover Page .............................................................................................................................................. 1

1.1 List of Figures ..................................................................................................................... 3

2 References ....................................................................................................................................... 4

3 Mission goal, science and technical objectives. ................................................................... 6

3.1 Mission Goal ............................................................................................................................ 6

3.2 Objectives ................................................................................................................................ 6

3.2.1 Science and Technical Objectives ............................................................................. 6

3.2.2 Specific objectives ......................................................................................................... 7

3.2.3 Optional Objectives ....................................................................................................... 8

4. Thermal Payload Analysis ......................................................................................................... 9

4.1 Thermal Payload Performance ........................................................................................... 9

4.2 Problems Encountered ......................................................................................................... 9

4.3 Lessons Learned .................................................................................................................... 9

4.4 Science Report ....................................................................................................................... 9

4.5 Technical Results ................................................................................................................. 10

5. Attitude Determination & Control System ......................................................................... 11

5.1 System Performance .......................................................................................................... 11

5.2 Problems Encountered ....................................................................................................... 14

5.3 Lessons Learned .................................................................................................................. 16

5.4 Technical and Science Results ......................................................................................... 17

6.0 Mechanical Payload Analysis: .............................................................................................. 19

6.1 Payload Performance: ........................................................................................................ 19

6.2 Problems Encountered: ..................................................................................................... 19

6.3 Lessons Learned: ............................................................................................................ 19

7 Electrical Payload Analysis ....................................................................................................... 22

7.1 Electrical System ................................................................................................................... 22

3

7.2 Data Measurement .............................................................................................................. 24

7.3 Electrical Issues ................................................................................................................... 26

8 Software Payload Analysis .............................................................................................................. 27

8.1 Problems Encountered................................................................................................................. 27

8.2 Lessons Learned........................................................................................................................... 27

9 Payload Member List ...................................................................................................................... 27

1.1 List of Figures

Figure 1 Experimental Data from our vacuum chamber ......................................................................... 10

Figure 2 Flight temperature data ......................................................................................................... 11

Figure 3, Detailed component diagram of Attitude Control System ........................................................ 12

Figure 4, Block diagram of the PID control system located in the dspic microcontroller. ........................ 13

Figure 5, HASP 2012 recorded attitude determination data in the form of angular rotations. ................ 18

Figure 6, HASP 2012 recorded attitude determination data in the form of angular rotations. ................ 18

Figure 7: Example of NX's structural simulation performed on one of the faces of the small cube. ......... 21

Figure 8: Example of NX's structural simulation performed on the entire small cube. ............................ 21

Figure 9: Example of NX's structural simulation performed on the entire payload. ................................ 22

Figure 10: Example of NX's structl simulation performed on the entire payload. ....... Error! Bookmark not

defined.

file:///C:/Users/Dmiralles/Desktop/dynamics/Dynamics%20Science%20report.docx%23_Toc343251763file:///C:/Users/Dmiralles/Desktop/dynamics/Dynamics%20Science%20report.docx%23_Toc343251764file:///C:/Users/Dmiralles/Desktop/dynamics/Dynamics%20Science%20report.docx%23_Toc343251765file:///C:/Users/Dmiralles/Desktop/dynamics/Dynamics%20Science%20report.docx%23_Toc343251766

4

2 References

[1] How Does an Accelerometer Work? (2011). Retrieved March 2011, from EzineArticles.com:

http://ezinearticles.com/?How-does-an-Accelerometer-Work?&id=285604

[2] Nave, C. (2010). Basic Rotational Quantities. Retrieved 03 10, 2011, from hyperphysics:

http://hyperphysics.phy-astr.gsu.edu/hbase/rotq.html

[3] Tawfik, H. (2009). A Glimpse at MEMS. Retrieved March 2011, from Knol:

http://knol.google.com/k/a-glimpse-at-mems#

[4] Digital Accelerometers. (2009). Retrieved March 2011, from Silicon Designs, Inc.:

http://www.silicondesigns.com/digital.html

[5] Pearson, C. (1999, July). How a Gyroscope Works. Retrieved March 2011, from

Gyroscopes.org: http://www.gyroscopes.org/how%5Chagwa4.pdf

[6] Newton's Laws of Motion. (2004, September 20). Retrieved March 2011, from MIT

OpenCourseWare: http://ocw.mit.edu/high-school/physics/systems-of-particles-

linearmomentum/impulse-and-momentum/8_01t_fall_2004_w03d1_class_06.pdf

[7] Comotti, D. (n.d.). ORIENTATION ESTIMATION BASED ON GAUSS-NEWTON METHOD

AND IMPLEMENTATION OF A QUATERNION COMPLEMENTARY FILTER. Retrieved from

dof-Orientation-Estimation: http://code.google.com/p/9dof-orientation-estimation/download

s/detail?name=GaussNewton_QuaternionComplemFilter_V13.pdf

[8] G. Welch, G. Bishop. (2006, July 24). An Introduction to the Kalman Filter. Retrieved June

20,2011, from The Kalman Filter: http://www.cs.unc.edu/~welch/media/pdf/kalman_intro.pdf

[9] Maxim Integrated Products. (2001, 11 21). TUTORIAL 2031 DC to DC Converter Tutorial.

Retrieved from http://www.maxim-ic.com/app-notes/index.mvp/id/2031

[10] National Intruments. (2011, Mar 18). Quadrature Encoder Measurements: How-To Guide.

Retrieved 2011, from http://www.ni.com

[11] Raygosa, R. (n.d.). Spanish tutorial for the Kalman Filter. Retrieved June 20, 2011, from

The KalmanFIlter:

http://www.cs.unc.edu/~welch/kalman/media/misc/RAYGOSA_KF_tutorial.zip

[12] M. Nieves, J. Quiñones. (2011, June 23). Theory For The Design of TigreSAT 2011 Attitude

Control System. Retrieved June 24, 2011, from CubeSAT Attitude Control System Team:

https://sites.google.com/site/cubesatacsteam/projectdescription/TheoryForTheDesignofTigreSAT

2011AttitudeControlSystem.pdf?attredirects=0&d=1

http://ezinearticles.com/?How-does-an-Accelerometer-Work?&id=285604http://hyperphysics.phy-astr.gsu.edu/hbase/rotq.htmlhttp://knol.google.com/k/a-glimpse-at-memshttp://www.silicondesigns.com/digital.htmlhttp://www.gyroscopes.org/how%5Chagwa4.pdfhttp://ocw.mit.edu/high-school/physics/systems-of-particles-linearmomentum/impulse-and-momentum/8_01t_fall_2004_w03d1_class_06.pdfhttp://ocw.mit.edu/high-school/physics/systems-of-particles-linearmomentum/impulse-and-momentum/8_01t_fall_2004_w03d1_class_06.pdfhttp://www.cs.unc.edu/~welch/media/pdf/kalman_intro.pdfhttp://www.maxim-ic.com/app-notes/index.mvp/id/2031http://www.ni.com/http://www.cs.unc.edu/~welch/kalman/media/misc/RAYGOSA_KF_tutorial.ziphttps://sites.google.com/site/cubesatacsteam/projectdescription/TheoryForTheDesignofTigreSAT2011AttitudeControlSystem.pdf?attredirects=0&d=1https://sites.google.com/site/cubesatacsteam/projectdescription/TheoryForTheDesignofTigreSAT2011AttitudeControlSystem.pdf?attredirects=0&d=1

5

[13] Colton, S. (2007). The Balance Filter. Retrieved March 2011, from

http://web.mit.edu/scolton/www/filter.pdf

[14] Glasse, P. C. (n.d.). An Introduction to the Use of Complementary Filters for Fusion of

Sensor Data. Retrieved 03 10, 2011, from glassercommunications:

http://www.glassercommunications.com/paul/samples/filters_for_fusion.pdf

[15] Nieves, M., Longoria, J., Quiñones, J., & Torres, A. (2011, May 24). MECN 3973

Aerospace Experience I Flight Readiness Review Document for the Accurate inertial

NavigaTion System (ANTS). Inter American University of Puerto Rico, Bayamón Campus,

Puerto Rico.

[16] Vo, H. et al., (2011, February 25). TigreSAT 2011 for ARIES. Retrieved July 25, 2011, from

TigreSAT 2011:

https://sites.google.com/site/tigresat2011/executivedocs/TIGRESAT_PDR_February_Revision_

V1.docx?attredirects=0&d=1

[17] VectorNav VN-100(T) User Manual. (2008). Retrieved December 5, 2011, from VectorNav

Technologies LLC: http://www.vectornav.com/Downloads/Support/UM001.pdf

[18] Kuipers, Jack B., “Quaternions and Rotation Sequences,” A Primer with Applications to

Orbits, aerospace, and Virtual Reality, Princeton University Press, New Jersey, 1999.

[19] Granger, D. (n.d.). LSU Science Space Group. Retrieved December 5, 2011, from Laspace

LSU: http://laspace.lsu.edu/hasp/Flightinfo-2011.php

[20] geomag.models@noaa.gov. (2010, March 6). NOAA national Geophysical Data Center.

Retrieved December 5, 2011, from The World magnetic

model:http://ngdc.noaa.gov/geomag/WMM/DoDWMM.shtml

[21] Wertz, J. R. (1980). Spacecraft Attitude Determination and Control . D. Reidel.

[22] Wertz, J. R. (1999). Space Mission Analysis and Design . Space Technology Library.

[23] Garmin Corporation. (n.d.). Retrieved November 16, 2010, from Garmin Website:

HYPERLINK "http://www8.garmin.com/aboutGPS/" 1

[24] Electronic Code of Federal Regulations. (2011, November 23). Retrieved November 24,

2011, from http://ecfr.gpoaccess.gov: http://ecfr.gpoaccess.gov/cgi/t/text/textidx?

c=ecfr;sid=fc43b2969257190ef4eea7d8d489e214;rgn=div5;view=text;node=14%3A4.0

.1.1.30;idno=14;cc=ecfr

[25] Bousoño Zavala, O. (2011). “Angular Position Control for the PR CubeSAT HASP”

http://web.mit.edu/scolton/www/filter.pdfhttp://www.glassercommunications.com/paul/samples/filters_for_fusion.pdfhttp://www.vectornav.com/Downloads/Support/UM001.pdfhttp://laspace.lsu.edu/hasp/Flightinfo-2011.php

6

3 Mission goal, science and technical objectives.

3.1 Mission Goal

The SWIM CubeSat is a 3U satellite that uses several subsystems to maintain stability and a pointing accuracy of 0.1 degree due to the science payload, WINCS.

The subsystem needed for this is an Attitude Determination and Control; however

this system is required to be proven as a reliable subsystem. Therefore an initial testing can be achieved through the HASP platform in a near space environment.

The development of this subsystem will also provide knowledge in aerospace

engineering to students of the Puerto Rico SWIM CubeSat project.

3.2 Objectives

These objectives are the measureable, simple and practical steps that are needed and will contribute to the fulfillment of specific goals.

3.2.1 Science and Technical Objectives

The following are several science objectives related to this experiment:

1. Increase the understanding of aerospace sciences specialized in CubeSat

technology.

2. Gain knowledge and experience in Attitude Determination such that it can

be applied for a CubeSat prototype.

3. Learn, test, and calibrate different sensors to measure the Earth’s

gravitational, and magnetic fields, also the Sun intensity, and the angular momentum of the ARIES Dynamics payload.

4. Learn about flight characteristic such as angular rate rotation, and orientation with respect to different reference frame.

5. Measure the internal temperature during flight to determine the potential

impact to the structure and the components of the structure.

6. Determine the position and the time of flight of the payload by using a

global

positioning system.

7. Acquire knowledge of post flight data analysis.

8. Obtained an appropriate quaternion representation through the Kalman filter processing.

7

9. Study the Earth magnetic field to obtain orientation of the CubeSat body

by using the Earth’s magnetic field as a reference frame.

10. Study the Angular velocity vs. time of the CubeSat during the HASP

flight.

11. Obtain knowledge of the acceleration of the CubeSat on the HASP

platform during the flight, thus permitting through mathematical equations

obtained the displacement and velocity by having an initial reference point.

12. Acquire information of internal temperature to obtain understanding of

the thermal behavior of the cube throughout the flight.

13. Obtain knowledge of the position and time of flight of the payload by

using a global positioning system.

14. Study the behavior of the Kalman filter with raw data of the HASP

payload.

15. Use the obtained data to gather knowledge about the attitude motion of

the payload and provide recommendations for future HASP and CubeSat

projects.

16. Learn about the thermal characteristic of the payload through the

environment that it will be exposed to through the flight.

17. Obtain knowledge of software implementation on a microcontroller.

3.2.2 Specific objectives

The specific objectives presented here are necessary to accomplish our

mission goal. These objectives will be completed during the development of the

prototype:

1. Apply astrodynamics, engineering, and programming concepts to design

software that will be capable of interpreting and using the data from the

Attitude Determination System, to stabilize the platform containing the

scientific package.

2. Design and connect the sensors located in a PC/104 board inside the body

frame of 10cm cube to determine the attitude motion of the payload. The

PC/104 is a CubeSat standard PCB.

3. Implement different attitude determination algorithms such as the Kalman

Filter and/or Newton Gauss method within the programming, to determine

the motion and orientation of the payload. The Newton Gauss method is not

8

essential although can increase the accuracy of the determination while

increase the computation complexity of the microprocessor.

4. Interpret the graphs obtained from the output of the different sensors as

well as the attitude of the vehicle. This data can be inserted in a Kalman filter

algorithm using MatLab to corroborate with the flight data a provide

5. Obtain the platform’s orientation with respect to the inertial reference

frame of the ARIES Dynamics project.

6. Reduce the attitude accuracy error of the payload to approximately 0.1

degrees. This can be performed with the Linear Covariance, which uses the

sensors to determine the uncertainty of error.

7. Develop the source code that will be used to control the platform

orientation in a closed loop system.

8. Implement a thermal system in the range of 0OC to 40oC to maintain the

components in the nominal operating temperature range.

9. Maintain the dimensions of the package within 15x 15 x 30 cm area.

10. Reorient the cube to a desired position using PID control.

11. Design, connect and develop a board capable of having both a Global

Positioning System and an Attitude Determination System while meeting the

PC/104 standards.

12. Comply with the CubeSat PC/104 mechanical standards.

13. Utilize the PC/104 Pumpkin bus for the communication between different

boards

14. Obtain internal temperature and pressure.

3.2.3 Optional Objectives

1. Determine the orientation of the payload with respect to the Sun using the

data obtained VI characteristic board.

2. Test the CINEMA solar panel circuit that will be used to acquire the

radiation measurements.

9

4. Thermal Payload Analysis

4.1 Thermal Payload Performance

The inside temperature of the payload is a very important parameter that has to be consider. If the inside temperatures reaches the critical temperatures the

effects on the payload are:

Damage on the circuit boards (possible shut down)

Thermal expansion on the structure (could cause bending stress,

structural failure)

The payload performance with respect to the thermal area was a success; the

temperatures inside the payload didn’t reach any critical conditions ( ), also no damage to the circuits board and structure was made with respect to the

temperature. The payload in regards to thermal did what was expected.

4.2 Problems Encountered

No significant problems were encounter with regards to the thermal design

and the inside temperature, although one of the thermostat did malfunction in flight reading and we know for a fact that this temperature reach is not possible (it could have been a possible reading error).

4.3 Lessons Learned

My personal experience with the HASP was a complete learning experience.

For the thermal it is interesting to note how the thermal environment in space

works. In example in space it is a complete vacuum so no convection is made, and

much more.

4.4 Science Report

Some basic assumptions were made:

Vacumm ambient

Heat conduction inside

Constante properties

Steady State

Note that steady state will tell us about the maximum and minimum

temperature the payload could reach.

Hot Case

Qemitted = 18.65 Watts and Tsur = 68.03°C.

10

Tin = 70.56°C.

Cold Case

Qemitted = 14.4 Watts and Tsur = 0.43°C.

Tin = 1.56°C.

These results are based on steady state analysis.

4.5 Technical Results

In the thermal design no used of “real” insulation was made because of the

powerful heater provided by the electrical engineers. This heater provided us with

an extreme internal heat load difference from 0W to 9W, that’s a big advantage for

the thermal design because when the temperature drop to 10 C the heater was

turned on and when it passed 20 C it was turned on, so basically you’ll have a

temperature distribution oscillating around 10 C or 20 C. The only problem could

had been caused by the sun, therefore we design for the surface to reflect most of

the suns radiation. The thermal design for the HASP Dynamic was a success the

proper amount of Kapton and a layer of mylar on the surface of the cube, gave a

Temperature distribution with respect to time inside the cube of almost a perfect

oscillation system.

Figure 1 Experimental Data from our vacuum chamber

0

10

20

30

40

50

60

70

0:00 0:28 0:57 1:26 1:55 2:24 2:52 3:21

Te

mp

era

ture

(C)

time(hour:minutes)

Inside Temperature vs. time

11

Figure 2 Flight temperature data

We can safely assume that the Hasp Dynamic could have another successful

flight with regards the thermal.

5. Attitude Determination & Control System

5.1 System Performance

The attitude and determination control sub-system (ADCS) of the payload

consisted of both electrical and mechanical components. As with most control

systems, ours could be separated into two different parts: sensors, controller, and

actuators. It’s purpose was to perform real-time, two-axis (pitch, yaw) attitude

control of the payload. Taking this into consideration we can classify the three

system components into electrical and mechanical aspects. The sensors and other

controller components are best considered from an electrical engineering point of

view, while the motors and moving parts are left to be seen from a mechanical

standpoint.

The attitude determination (sensors) part of our system relied on

measurements taken from a magnetometer, accelerometer, and gyroscope.

0

10

20

30

40

50

60

70

0 1 2 3 4 5 6 7 8 9 10

Tem

pe

ratu

re (C

)

time (hour)

Inside Temperature vs. time

12

Together these can be found in IMU’s (inertial measurement units) as is the case

for our system. The measurements taken from the IMU (Razor 9dof IMU) were

filtered through Kalman and other complimentary filters to calculate the angular

velocity, near magnetic fields, and acceleration of the system.

These filters were implemented taking into account the limitations and

performance of the separate IMU components and were of paramount importance

when trying to minimize gyro drifting and fusing separate sensor signals. To further

reduce measurement uncertainty, calculations were performed in the form of

quaternions and only given rotation angle format when stored for user processing.

HASP

30 V@.5A

Primary DC to DC

Converter

30V to 9V

DC to DC Coverter

9V to 5V

Secondary DC to

DC Converter

9V to 3.3V

DC to DC Co

9V to 5V

dspic33FJ256GP710AMicrocontroller

Mosfet Switch

Mosfet Switch

H-Bridge Pitch Motor Control with

Encoder

H-Bridge Yaw Motor Control w Encoder

Real Time Clock

Quadrature Decoder

ADS Board Microcontroller

PC/104

Legend___________ Power

_______ Communication

Legend___________ Power

_______ Communication

SDcard

Figure 3, Detailed component diagram of Attitude Control System

The attitude control (actuator) part of our system was simpler in that it relies

only on two inputs: the desired pitch and yaw angles. The actuators were simply

two custom Maxon DC motors connected via timing belts to separate gears fixed to

the structure. Each motor was responsible for a single rotation direction. This

configuration has the advantage that the timing belts were of such length so that

the structure remained rigid when the motors were stationary. That is to say, we

only had to worry about HASP platform attitude since the satellite would be

stationary with respect to it.

The final part of the system is of course, the controller itself. While this does

not strictly fall into a mechanical or electronic classification as suggested above, it

13

needs to be so seamlessly a part of both that it would be harder to describe is as

anything else. For simplicity, the controller of choice was a PID controller. The

corresponding gains of the system (proportional, integral, and differential) were

determined after analyzing the dynamics of the system using methods such as pole

placement and Zeigler-Nichols tuning. Armed with measurements such as motor

constants, rotational moments of inertia, and others it was also possible to

determine the system’s transfer functions:

G(s)p =

=

=

G(s)y =

=

=

Where subscripts p and y indicate pitch and yaw respectively. The basic

equation and diagram for the chosen standard PID controller are shown below.

Figure 4, Block diagram of the PID control system located in the dspic microcontroller.

Using SIMULINK’s PID design tools it was possible to determine (using the

system’s EOM (equations of motions)) the ideal values for the PID were determined

and then compared to the calculated ones. As expected the results were similar but

Erro

r

output Trajectory -

+ +

+

Disturbances

Output Transducer

Proportional gain

Integral gain

Derivative gain

)(teKP

t

i deK0

)(

dt

tdeKD

)(

Σ Σ

14

some discrepancies existed. These were resolved by simply testing the system and

then tweaking the gain values back and forth between the expected values until the

desired response was achieved.

The resulting controller gave an output with a steady state error of

approximately +/- 1° which is remarkable performance. While theoretically there

should be no error as long there is integral gain present it should be considered

that no sensor or filter is perfect, therefore there is bound to be some error.

5.2 Problems Encountered

While the performance described above was indeed achieved, thus meeting

the set accuracy requirements (+/- 5°). Some unforeseen complications occurred

between this time and the time of flight.

As is often the case with these situations, the exact cause of mistakes and

shortcomings is not completely certain. For this reason, this section will be outlined

such as to state what went wrong, the effects of this mistake, and then possible

causes. In order of occurrence:

- H-bridges Stopped Working:

The h-bridge is an electrical component which main purpose in our system is

to allow a DC motor to rotate in either direction using logic voltages. That is, it

reverses the current flow through the motor. At different points both h-bridges

malfunctioned rendering the control system unusable as the system could only

compensate for disturbances in a single direction. The reason for this mishap is not

clear but is most likely due to carelessness while handling the payload. Although

since both components failed, it is possible there was a flaw in the soldering or

circuit design. It should be noted that the yaw h-bridge failed two months before

the pitch h-bridge did.

- Payload Stalled or did not Rotate:

The second problem occurred after the first h-bridge failed. This time the

payload rotated but chose an arbitrary 0° position every time it was started and

while it did respond to external disturbances and changes in position it did so

erratically. After some analysis it was determined that a new Kalman filter had to

be implemented and that the PID controller gains had to be recalculated. Several

attempts were made to fix this but though the attitude correction did improve it did

15

not regain the same accuracy as before. Furthermore after a few trials the problem

started to repeat itself.

The reason behind this problem was never clear. Since the student who

designed the initial control system was not available during the summer period

there was little the current team could do other than to implement a different

control system. The result was a simple circumstantial feedback loop system with

several nested conditions. The idea was simple and effective but limited. The

system was able to compensate for disturbances in noticeably larger steps

depending in the magnitude of the offset from its zero point, however, these steps

were marked by noticeable pauses instead of a fluid motion as with the PID

controller. On the other hand, with this system there was zero overshoot.

- Payload Rotated Continuously and was Irresponsive:

During the thermal testing at the Balloon Facility in Texas the payload began

to exhibit some unusual behavior after it had spent more than an hour inside the

vacuum chamber. In these occasions the payload would begin to spin continuously

in one direction and become unresponsive. At the time it was thought to be a

problem with the IMU board but later it became clear that the implemented control

code did not account for cumulative gyroscope drift. This caused this error to build

up and eventually the system became unable to compensate and exhibited the

discussed reaction. Later, the pitch control h-bridge was determined to be

functioning intermittently.

- No Live HASP Communications:

During the HASP flight no live data was received. There is no way of verifying

the cause of this but it could range anywhere from an electrical subsystem failure

to damage acquired during shipment and handling. For more details see below.

- Corrupted Attitude Data:

The biggest problem of all was none other than data corruption. Upon the

return of our payload we attempted to retrieve the flight data stored in the system

SD card but upon further analysis (shown in the next section) we discovered that all

data except for the temperature reading storage was corrupt and completely

unreadable in places. In addition, it appears that the attitude determination system

stopped working (or at least the data storage process) and restarted several time

during flight.

16

The evidence above suggests that the problem lies not in the control software

but rather in the control system board. Particularly the DC to DC 9V to 5V

regulator. In Figure 1 this it can be seen that this component powers not only the

ADS microcontroller but also the yaw rotation h-bridge and the real time clock. All

of which were parts that experienced problems. In fact, a faulty power regulator

could have been responsible for damaging both the h-bridge IC and the attitude

data by continuously cycling power to the ADS microcontroller (which the data

shows happened at least 30 times during flight).

5.3 Lessons Learned

Of the many lessons learned, perhaps the most important from a controls

system perspective is to know your sensors. Many of the problems could have been

properly diagnosed if not for the fact that many a times we did not know in which

system in particular the problem lay. The issue with the gyroscope drift clearly

shows how ignorance or lack of information on a small matter can jeopardize the

entire project.

Furthermore, another thing of great importance when working on a team

containing multiple members of different concentrations and specialties is

communication. What may be perfectly obvious to one person could be confusing or

completely unknown to another. This weighs even more when considering that

some members of the group were forced to leave and were replaced with others

that while skillful were completely new to the project.

It is no doubt hard to find a person, much more a student, who is

knowledgeable in enough areas to do an entire task unassisted by peers with

different knowledge and skills, however, it should be a goal while a member of a

team to become familiar with as many areas as possible. This way, while carrying

out your own assignment it will be possible for you to understand when your

decisions affect other areas of a project and whether those decision are correct. In

our particular case, if those working on control had been more familiar with the

basics of microcontroller programming and sensor integrated circuits and

electronics, the previous section may have been much shorter than it is.

That being said, it should be noted that as a student project the experience

was by no means a failure. Despite the mission’s shortcomings some data was

retrieved safely and other subsystems worked correctly and successfully. It has

been an invaluable opportunity that will undoubtedly yield fruitful experience for all

involved.

17

5.4 Technical and Science Results

As stated above, since the attitude data was corrupted there is very little in

terms of results in the attitude control area. The following figures show how the

recorded data compares with the same type of data collected in a laboratory test of

the same system. As can be see, the collected data does not even measure in the

same order of magnitude as the lab data. In addition the sampling time labels were

also corrupted, giving the data the ‘blank’ appearance displayed in the Figure 5.

18

Figure 5, HASP 2012 recorded attitude determination data in the form of angular rotations.

Figure 6, HASP 2012 recorded attitude determination data in the form of angular rotations.

19

6.0 Mechanical Payload Analysis:

6.1 Payload Performance:

In terms of performance the CubeSat (structurally speaking) performed

almost perfectly. There were no major flaws in the design and it suffered no cracks

or serious deflections (the only one being a small twist in the shaft which did not

cause the system to fail).

The small cube rotated without crashing into any of the other parts like it did

before, the motors didn’t stop working nor did the rest of the components freeze

over and break. Speaking in terms of functionality, the project was a success.

6.2 Problems Encountered:

The biggest problems encountered that dealt with the mechanical part of the

design were in the manufacturing part of the project. The university did not have

the tools necessary to build the CubeSat so we found ourselves in need to employ

outside resources, which resulted in many inconveniences. These inconveniences

were of economical and temporal nature. We had to find a machine shop, either

industrial or from another university, which would do the labor cheaply and quickly.

At first we thought we were in the clear when we got some students from the

same university who were working on their capstone project to do the work for us

for free but the machines they were using broke down and half of the payload’s

parts were incomplete. With the little time we had we searched desperately for a

machine shop that would get the things done before we ran out of time. We

enlisted the help of a student from the University of Florida to get the job done.

6.3 Lessons Learned:

Throughout the entire project our team acquired and mastered many new

skills. Many of our members were new to designing and manufacturing so taking

part in all the processes that involved applying their knowledge to the building of

the CubeSat’s mechanisms proved to be both a challenge and a good experience.

Improving upon the previous design of the CubeSat was not very difficult

since it only amounted to fixing a few minor details like the cube not being able to

turn completely. The real challenge lied with the manufacturing part of the project

20

as described in the previous section. Thanks to that experience our team gained

valuable knowledge and experience that will serve us well in future projects of the

same nature.

It can be said that the team learned how to use CAD software better than

before and learned how to run structural and thermal simulations through them.

Also, we put our interpersonal skills to the test since, again, most of us haven’t

worked in projects such as this before and haven’t really seen the limits of our

skills. All in all our members grew, not only as a team, but also as individuals

seeking to become better professionals in the fields of science and engineering.

6.4 Science and Technical Results:

In a static analysis the best method to determine if the structure is going to

fail by yield is applying the Distortional Energy Theory best known as the Von-

Misses theory or the Maximum Shear Stress Theory. To be less conservative we

used the Von-Misses theory (have in mind that this is the most used in common

practice). This theory is based on the idea that yielding occurs when the distortion

strain energy reaches or exceeds the distortion strain energy for yield in simple

tension or compression, that’s why it is only used for ductile materials. The ABS+

and Aluminum 6061 T6 have so they are considered ductile materials ergo

the Von-Misses theory is applicable.

The previous equation is the Von-Misses equation for a 3-dimension analysis,

for simplification the analysis is made in a 2 dimension so the equation boils down

to:

Where is the Von-Mises Stress, is the stress in the x component, is

the stress in the y component and is the shear stress applied in the y

component where the stress in x takes place.

After applying the above equation for a simple analytical approach and later

corroborating the results by performing numerical solutions with a CAD program

(NX) we were certain that the structural integrity of the payload wasn’t an issue.

But as the flight proved, you cannot always rely completely in software simulations.

During the flight, the payload suffered a minor deflection in one of its shafts and,

although it resulted in no problems, it showed that more serious tests than the ones

performed are needed to make sure everything will go as planned.

21

The following figures are some examples of our simulations using the CAD

program NX:

Figure 7: Example of NX's structural simulation performed on one of the faces of the small cube.

Figure 8: Example of NX's structural simulation performed on the entire small cube.

22

7 Electrical Payload Analysis

7.1 Electrical System

The electrical system is divided in three main parts; ADS board, Heater board

and ACS board. In the ADS board is where the principal power conditioning

components are located. The primary DC/DC converter converts the 30VDC from

the HASP platform to 9V. Then the secondary converters convert those 9V into 5V

and 3.3V. These voltages are distributed into the different components and boards

in the payload. For example, the H-Bridge Yaw motor control, Real Time Clock and

ADS microcontroller are been fed by one of these converters. The other

two secondary converters supply voltage to the H-bridge Pitch control and

dspic microcontroller. Figure x shows in more detail the power conditioning and

distribution in the ADS board.

Figure 9: Example of NX's structural simulation performed on the entire payload.

23

Figure 10 Attitude control system and Flight Board system diagram

The components used on the heater board are two 2F7455 MOSFET, one

dspic microcontroller and three hi-power resistor, which work as the component in

charge of transforming the electrical power into heat. The temperature on the

payload is controlled on this board and start with the microcontroller which is

programmed to send a signal (high) to the MOSFETs, we named this time interval

as the “ON period” of the heater. The MOSFETs are used as a simple switch which

allows the pass of current whenever the dspic send the signal. This dspic is

programmed to be in the “ON period” when the internal temperature on the

payload reaches 0 oC or below. Figure x shows the schematic design created in

Eagle of the heater board.

24

Figure x:

Figure 11 Heater Schematic Design

7.2 Data Measurement

The measurement of the voltage supply by the HASP platform was taken.

This is important to our team because the selection of the Integrated Circuits is

based on those voltage values. Figure x shows the voltage supply during the

complete flight. This graphs shows that the average voltage supply remains in 29V,

not getting higher that 31V or lower than 28V. Also the data shows approximately a

dozen times in where the power was reset. This can lead to a problem if the interval

of time is too long, because the heater could not turn on in case of any extreme

temperature.

25

Figure x

Figure 12: Voltage supply during flight

The HASP Platform provides different amount of power consumption

depending on the payload size (big or small). In our case (small payload) the

platform provides us with 30VDC @ 0.5A, this means that we cannot exceed 15W of

power consumption. In the graph shown in Figure x the data related to the power

draw was taken. The power draws state most of the time at ~4W, with a high peak

of 12.2W for a short interval of time. This peak happened when the heater turn on

(low temperature). This data demonstrate that the electrical part regarding the

heater was successful.

-5

0

5

10

15

20

25

30

35

"09/

01/1

2 13

:13:

27

"09/

01/1

2 13

:30:

01

"09/

01/1

2 13

:46:

11

"09/

01/1

2 14

:02:

39

"09/

01/1

2 14

:19:

00

"09/

01/1

2 14

:35:

41

"09/

01/1

2 14

:52:

05

"09/

01/1

2 15

:08:

15

"09/

01/1

2 15

:24:

25

"09/

01/1

2 15

:40:

34

"09/

01/1

2 15

:56:

44

"09/

01/1

2 16

:12:

54

"09/

01/1

2 16

:29:

04

"09/

01/1

2 16

:45:

13

"09/

01/1

2 17

:01:

23

"09/

01/1

2 17

:17:

33

"09/

01/1

2 17

:33:

43

"09/

01/1

2 18

:15:

44

"09/

01/1

2 18

:55:

30

"09/

01/1

2 19

:11:

48

"09/

01/1

2 19

:27:

58

"09/

01/1

2 19

:44:

21

"09/

01/1

2 20

:01:

07

"09/

01/1

2 20

:17:

30

"09/

01/1

2 20

:40:

56

"09/

01/1

2 20

:57:

08

"09/

01/1

2 21

:13:

39

"09/

01/1

2 21

:31:

21

"09/

01/1

2 21

:47:

34

"09/

01/1

2 22

:05:

25

"09/

01/1

2 22

:21:

34

"09/

01/1

2 22

:37:

44

"09/

01/1

2 22

:53:

55

Vo

ltag

e (V

olt

s)

Voltage - Dynamic Payload

26

Figure 13: Power consumption during flight

7.3 Electrical Issues

As soon as the payload arrives to Puerto Rico after flight, electrical tests were

performed to all the boards in the payload. In those tests we found that the heater,

ADS and ACS board worked as expected, with the exception of one DC/DC

converter on the ADS board. This converter fed the H-bridge Yaw motor and the

ADS microcontroller. At first we thought that the failure in this converter was not a

big issue, but after analyzing the SD card data we found that it was corrupted (with

the exception of the temperature data) due to this problem. A single component

like this cause major data loss of the experiment, e.g. the yaw, rolls and pitch data.

The cause of this failure was because the external capacitor of the dc/dc

converter blew up. Several reasons could be the cause of this problem like;

Electromagnetic Interference (EMI), short circuit and/or vibration problems. We do

not had enough understanding on the EMI field and that’s why we do not perform

any simulation nor implementation regarding the EMI control. Some vibration tests

were performed on the payload before the flight (which were success), but maybe

more deeply vibrations tests were needed to discard any possibility.

The work ahead is mainly oriented to understand and perform all the

necessary tests and simulations to ensure a proper performance during flight. At

least electromagnetic interference simulations have to be under our consideration

for the next payload project.

0

2

4

6

8

10

12

14

"09/

01/1

2 13

:13:

27

"09/

01/1

2 13

:30:

01

"09/

01/1

2 13

:46:

11

"09/

01/1

2 14

:02:

39

"09/

01/1

2 14

:19:

00

"09/

01/1

2 14

:35:

41

"09/

01/1

2 14

:52:

05

"09/

01/1

2 15

:08:

15

"09/

01/1

2 15

:24:

25

"09/

01/1

2 15

:40:

34

"09/

01/1

2 15

:56:

44

"09/

01/1

2 16

:12:

54

"09/

01/1

2 16

:29:

04

"09/

01/1

2 16

:45:

13

"09/

01/1

2 17

:01:

23

"09/

01/1

2 17

:17:

33

"09/

01/1

2 17

:33:

43

"09/

01/1

2 18

:15:

44

"09/

01/1

2 18

:55:

30

"09/

01/1

2 19

:11:

48

"09/

01/1

2 19

:27:

58

"09/

01/1

2 19

:44:

21

"09/

01/1

2 20

:01:

07

"09/

01/1

2 20

:17:

30

"09/

01/1

2 20

:40:

56

"09/

01/1

2 20

:57:

08

"09/

01/1

2 21

:13:

39

"09/

01/1

2 21

:31:

21

"09/

01/1

2 21

:47:

34

"09/

01/1

2 22

:05:

25

"09/

01/1

2 22

:21:

34

"09/

01/1

2 22

:37:

44

"09/

01/1

2 22

:53:

55

Po

we

r (W

atts

)

Power - Dynamic Payload

27

8 Software Payload Analysis

8.1 Problems Encountered

After the integration process at Palestine, TX the payload suffered several

problems and we take it back home for fixing it and to let it ready for the flight in

New Mexico. During the process of reviewing the payload the serial communication

line TX and RX was inverted disabling the possibility of establishing the serial

communication channel with the payload. That error limited our possibility of any

kind of troubleshooting with the device since the payload did not sent the

housekeeping data.

In addition when the data analysis was produced we manage to see that

there were not any force reset state with the software, since all the data logging

was continuous and was only interrupted when the payload was powered on and

off. This was checked using the data logging of the HASP platform, specially on the

files that showed the current and voltage of the payload.

Even though the software was running as expected we did notice that our

ADCS data was in part corrupted so our team believe that the IMU was damaged

during the flight but its data logging was performed as expected.

8.2 Lessons Learned

The most important lesson learned was that space environment is really

hostile and that it is never enough all the little details that could help to prevent a

malfunction. For that reason our team considered that for future flights we could

create a more self intelligent system capable of performing self-corrections if it is

flown without serial communications. So ,let say that our MCU detects that the IMU

data is being normally received and logged but that its content is corrupted. So our

system could consider that kind of event as a malfunction and perform a reset in

order to see if the problem was properly fixed.

In addition we learned that we have to double check all the steps that we do

until we are sure it is properly working. This previous statement is more focused to

our serial communications mistake and the consequences of the action.

9 Payload Member List

28

Figure 14 List of the Personnel involved in the AIES-Dynamics payload.

The previous table show all the member that take action in the creation of this

payload. As listed some of the member who hardly work on the project are already

employed thanks to the training an the experience that platform like HASP offered

to our group.